own/metal
← LLM Runners tool · /LOCALA

LocalAI

Drop-in OpenAI-compatible API for local models.

// github

★ 46.6k

last commit · today

heavy CPU only MIT

// readme · what it is

LocalAI exposes an OpenAI-compatible REST API backed by llama.cpp, whisper.cpp, Stable Diffusion and more. If you already have code that calls `client.chat.completions.create(...)`, you can point it at LocalAI and get the same shape back. Useful when you want one endpoint that handles chat, embeddings, transcription and image generation.

// deploy notes

Heavier than Ollama because it bundles multiple backends. Worth it if you want one API for chat + embeddings + image.

[ ALTERNATIVE TO ]